Enhancing high-temperature energy density of dielectric composites by tailoring molecular semiconductors

Abstract

To overcome the challenge where the exponential increase in leakage current of aromatic dielectric polymers under high-temperature and high-electric-field conditions leads to a mutual constraint between capacitive performance and thermal stability, in this study, diketopyrrolopyrrole derivatives (DPP-X, X = Cl, Ph, and O) with different electron-donating groups were doped into the PEI matrix. Based on the differences in the electron-donating ability of these electron-donating groups, the depth of charge carrier (electron and hole) traps in the PEI-DPP-X (X = Cl, Ph, and O) composite materials, as well as the strength of hydrogen bonding and electrostatic interactions between DPP-X (X = Cl, Ph, and O) and the PEI matrix, can be optimized—ultimately achieving an improvement in their high-temperature capacitive energy storage performance. As a result, DPP-O, with the strongest electron-donating ability, constructs deeper hole traps. At 200 °C, compared with pure PEI, the leakage current of PEI-DPP-O (0.2 wt%) is reduced by 3.37 times. Meanwhile, due to the strong electron-donating groups of DPP-O, it forms a more robust physical crosslinking network with the PEI matrix through hydrogen bonding and electrostatic interactions, enhancing the mechanical strength and breakdown strength (E_b). This synergistic regulation mechanism achieves the joint enhancement of electrical insulation and heat resistance of composites, enabling PEI-DPP-O-0.2 wt% composites to achieve excellent energy storage densities of 5.64 J cm⁻³ and 3.98 J cm⁻³ at 150 °C and 200 °C (η = 90%), respectively, outperforming lots of reported dielectric composites. In addition, the excellent cycling stability, the ability to prepare large-area high-quality uniform thin films, and the ultra-low filler loading characteristics further confirm the potential application of dielectric composites through molecular structure level regulation under extreme conditions.

---

1. Introduction

Electrostatic capacitors possess characteristics such as high-power density, fast charge–discharge rates, and high reliability and are widely used in pulsed power systems and electronic systems.^1–3 Traditionally, the dielectrics for electrostatic capacitors are mainly divided into two categories: ceramics and polymers. Polymer dielectrics have become the preferred choice for electrostatic capacitor applications due to their excellent processability, high breakdown strength, light weight, low cost, and strong charge–discharge recyclability.^4,5 In recent years, with the development of aerospace, wind power generation, underground oil and gas exploration and other fields, higher requirements have been put forward for the high-temperature tolerance of polymer dielectrics.^6,7 Therefore, aromatic polymer dielectrics with high glass transition temperature (T_g) have been developed for high-temperature energy storage. For example, polyetherimide (PEI) exhibits high thermal stability with a T_g as high as 217–270 °C, but its abundant conjugated structures lead to an exponential increase in leakage current with the increase of temperature and electric fields. This means that the energy storage performance and thermal stability of these aromatic polymers are often mutually exclusive under high-temperature and high-electric-field conditions.^8–10 Therefore, suppressing the growth of leakage current in polymer dielectrics under high temperature and high electric fields and developing a new generation of polymer dielectrics with high temperature resistance (≥150 °C), high energy storage capacity, and high efficiency (η ≥ 90%) are keys to solving the above problems.^11–13

Introducing molecular semiconductors into aromatic engineering polymers with high thermal stability represents one of the important strategies for preparing polymer dielectrics that concurrently feature high temperature resistance and high energy storage characteristics.^14–18 Due to their strong electron/hole affinity, molecular semiconductors are receiving increasing attention in the field of thin film capacitors.¹⁹ Molecular semiconductors have lower LUMO and higher HOMO energy levels than polymer matrices, which can serve as carrier trap sites to capture carriers in molecular semiconductor/polymer composites, thereby reducing leakage current and improving energy storage density.^20–22 The depth of traps in molecular semiconductor/polymer composites depends on the difference in their LUMO/HOMO energy levels. Therefore, trap depth can be regulated by changing the band structure of molecular semiconductors or polymer matrices. For example, Wang et al.²³ reported a general method for controlling charge trap energy levels in all organic polymer composites through organic semiconductor substituent engineering. They focused on studying the influence of the structure and properties of semiconductor molecules on the capacitance performance of composite materials. Furthermore, methods such as introducing inorganic fillers or crosslinking into polymer matrices can also enhance the energy storage performance of polymer dielectrics at high temperatures, but they are accompanied by issues like sacrificed flexibility, interfacial incompatibility, and complex preparation processes.^24–27 Shen et al.⁵ introduced 3D rigid aromatic molecules into aromatic polyimides, forming a physical crosslinking network through electrostatic interactions between their oppositely charged phenyl groups. The dense physical crosslinking networks strengthen polyimides to boost the breakdown strength (E_b), while the aromatic molecules trap charge carriers to suppress losses, enabling this strategy to combine the advantages of inorganic doping and crosslinking. These research results demonstrate that molecular-scale structural design (including the regulation of electronic structures of semiconductor molecules and the design of intermolecular interactions) is an effective approach for addressing the contradiction between high-temperature energy storage performance and high-temperature resistance of polymer dielectrics.

In this work, a series of donor–acceptor–donor (D–A–D) semiconductor molecules—diketopyrrolopyrrole derivatives (DPP-X; X = Cl, Ph, and O) with a tailored electron-donating group structure—to precisely modulate their energy band structures are proposed. These DPP-X (X = Cl, Ph, and O) were doped into the polyetherimide (PEI) matrix. Based on the differences in the electron-donating ability of the electron-donating groups in DPP-X (X = Cl, Ph, and O), the depth of carrier (electron and hole) traps formed in the PEI-DPP-X (X = Cl, Ph, O) composites and the strength of hydrogen bonding and electrostatic interactions between DPP-X (X = Cl, Ph, and O) and the PEI matrix can be optimized. Among them, the electron-donating group of DPP-O has the strongest electron-donating ability, forming deeper hole traps in the PEI-DPP-O-0.2 wt% composite, which reduces the leakage current of the composite by 3.37 times compared with that of the pure PEI film at 200 °C. Moreover, due to the strong electron-donating groups of DPP-O, it more readily forms a robust physical crosslinking network with the PEI matrix via hydrogen bonding and electrostatic interactions, significantly enhancing the mechanical strength and breakdown strength (E_b) of the dielectrics. At room temperature, 150 °C, and 200 °C, the E_b of PEI-DPP-O-0.2 wt% is 1.17 times, 1.11 times, and 1.14 times that of pure PEI, respectively. Furthermore, the strong electron-donating ability of DPP-O enhances the dipole moment and interfacial charge accumulation of the composite, leading to an increase in the dielectric constant by improving the dipole and interfacial polarization intensities. Therefore, at room temperature, the U_d of PEI-DPP-O-0.2 wt% reached 7.38 J cm⁻³, which is 1.92 times higher than that of pure PEI (3.85 J cm⁻³), while maintaining an η of 90%. Most importantly, even at high temperatures, PEI-DPP-O-0.2 wt% can achieve excellent energy storage densities of 5.64 J cm⁻³ and 3.98 J cm⁻³ at 150 °C and 200 °C, with an η of 90%, which is superior to most reported high-temperature energy storage polymers. Furthermore, we have confirmed that the PEI-DPP-O-0.2 wt% composite material can be used to prepare large-area uniform films and exhibits excellent cycling stability at 200 °C. In summary, this study systematically explored the influence mechanism of electron donating groups on the electron cloud distribution, band structure, and interface interactions of DPP derivatives at the molecular level. Furthermore, it revealed the regulation rules of electron donating groups on core performance parameters such as the dielectric constant, breakdown strength, and energy storage density of PEI-DPP-X composites, providing a new theoretical basis for the optimization design of high-temperature energy storage performance of polymer dielectric materials.

2. Experimental

2.1. Materials

4,4′-(4,4′-Isopropylidenediphenoxy)bis(phthalic anhydride), m-phenylenediamine, 3,6-diphenyl-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione, and 3,6-bis(4-chlorophenyl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione were purchased from TCI Shanghai. 3,6-Di(furan-2-yl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione was purchased from Shanghai Macklin Biochemical Technology Co., Ltd. 1-Methyl-2-pyrrolidone (NMP, A.R., >99.0%) was purchased from Sinopharm Group Chemical Reagent Co., China. All raw materials were used without further purification.

2.2. Preparation of PEI-DPP-X (X = Cl, Ph, and O) composites

The solution casting method is used to prepare PEI-DPP-X composite dielectric films. Firstly, m-phenylenediamine (MPD) and 4,4′-(4,4′-isopropylidenediphenoxy)bis(phthalic anhydride) (BPADA) (a molar ratio of 1:1) were added to a 20 mL sample bottle, and a certain amount of solvent N-methyl-2-pyrrolidinone (NMP) was added. The mixed solution was stirred at room temperature for 2–3 h to obtain a colorless transparent viscous poly(amide acid) (PAA) precursor. Then, the NMP solution of DPP-X (X = Cl, Ph, and O, 2 mg mL⁻¹) was sonicated for 20 minutes and added to the polymer matrix in a certain volume to maintain the mass fraction of the polymer at 13%. Among them, the doping ratio of the small molecule DPP-X (X = Cl, Ph, and O) is set to 0.2 wt%, which is the optimal value determined through preliminary investigation. As shown in Fig. S1, with the increase in the doping ratio of the small molecule DPP-O, both the energy storage density (U_d) and breakdown strength (E_b) of the PEI-DPP-O composite film show a trend of first increasing and then decreasing; when the doping ratio of DPP-O is 0.2 wt%, the U_d and E_b of the PEI-DPP-O-0.2 wt% composite film reach their optimal values. Therefore, to investigate the effect of the structure of electron-donating groups in DPP-X on the high-temperature energy storage performance of composite materials by controlling a single variable, this study selects 0.2 wt% as the optimal doping ratio. The mixture was stirred at room temperature for 12 h to evenly disperse DPP derivatives in the polymer matrix. Subsequently, in a 70 °C blast drying oven, the PAA solution with DPP-X (X = Cl, Ph, and O) was coated onto clean glass plates with a 200 μm scraper, and then thermally imidized by heating at 80 °C for 3 h, 100 °C for 1 h, 150 °C for 1 h, 200 °C for 1 h, 250 °C for 1 h, and 300 °C for 1 h in a vacuum oven. After complete thermal imidization, the temperature of the oven was allowed to cool naturally to room temperature; the glass plates were soaked in deionized water, the film from the glass plate was peeled off, and the unreacted impurities and by-products were removed. Then, the film was dried in an 80 °C vacuum oven for 24 hours to remove moisture and the final PEI-DPP-X (X = Cl, Ph, and O) composite film was obtained. The preparation process of pristine PEI films was the same as that of blended films. According to the different electron-donating groups of DPP derivatives, the obtained films are represented as PEI-DPP-Cl-0.2 wt%, PEI-DPP-Ph-0.2 wt%, and PEI-DPP-O-0.2 wt%.

2.3. Material characterization and electrical measurement

The characteristic functional groups of polymer dielectric films were studied using a Fourier transform infrared (FTIR) spectrometer (Is-50, Thermo Fisher) in the range of 500–4000 cm⁻¹. Ultraviolet-visible (UV-vis) absorption spectra were obtained using an UV3600IPLUS spectrometer (Shimadzu). The Young's modulus was obtained using Nano-indentation tests (KLA G200X, USA). An X-ray diffractometer (XRD, Smartlab SE, Rigaku) equipped with a Cu Kα source with a wavelength of 1.54 Å was used to characterize the condensed structure of polymer films. Differential scanning calorimetry (DSC; DSC-25, America) was used to characterize the thermodynamic properties of polymer films, with a heating and cooling rate of 3 °C min⁻¹. Before the electrical measurements, Au electrodes of 2 or 8 mm in diameter were deposited on both sides of the films using a JS3S-60 magnetron sputtering coater with a power of 120 W and a sputtering time of 360 s. An impedance analyzer (E4990A) was used to study the effect of the frequency (1 kHz to 10 MHz) on the dielectric constant of polymer films at room temperature. A dielectric temperature spectrum measurement system (DMS-500, China) was used to study the effect of temperature on the dielectric constant from 30 °C to 200 °C. Electric displacement–electric field curves (D–E loop) under 100 Hz and at RT, 100 °C and 200 °C were characterized by a ferroelectric polarization loop and dielectric breakdown test system (PolyK Technologies State College, PA, USA). The leakage current and the cycling stability of polymer dielectric films were characterized using a TF Analyzer 3000 ferroelectric polarization tester (aixACT, Germany). A rapid charge–discharge test was conducted using a DCQ-20A charge measurement system (PolyK Technologies LLC., USA) with a resistance of 10 kΩ.

3. Results and discussion

3.1. Structural design and molecular trap regulation of DPP derivatives

As shown in Fig. 1a, three DPP derivatives with D–A–D structure contain different electron-donating groups, which are chlorobenzene, benzene and furan in turn, and are correspondingly named DPP-Cl, DPP-Ph and DPP-O. The electron donating group changes the electron distribution of DPP molecules through an inductive effect or a conjugation effect, directly affecting the molecular orbital energy levels (HOMO/LUMO) and energy gaps. This mechanism has been validated through the combination of DFT calculations and organic chemistry theory.

Fig. 1 (a) Trap sites created between DPP-X (X = Cl, Ph, and O) and the polymer PEI matrix. (b) The total number of electrons in the acceptor groups of DPP-X (X = Cl, Ph, and O). (c)–(e) Electrostatic potential distribution and area percentage in each electrostatic potential range of DPP-Cl, DPP-Ph, and DPP-O.

Here, GaussView 6.1.1 for Windows is used for modeling, observing the spatial distribution of electrostatic potential as well as characterizing the LUMO and HOMO, while Gaussian16 C.01 for Linux is adopted for geometric structure optimization at the b3lyp/6-31g (d, p) level without symmetry constraints in all calculations. DFT calculations have also verified that the HOMO energy level of DPP-O is higher than that of DPP-Ph, and the HOMO energy level of DPP-Cl is the lowest among the three (Fig. 1a). From the perspective of organic chemistry, furan is a functional group with strong electron-donating ability. The oxygen atom in the furan ring has lone pair electrons, which can donate electrons to the conjugated system of the DPP molecule through the conjugative effect. This electron-donating effect increases the overall electron cloud density of the molecule. According to molecular orbital theory, the entry of more electrons into the molecular orbitals causes the HOMO energy level to rise. Therefore, when the electron-donating group of DPP is furan, the HOMO energy level is relatively high. The electron cloud distribution of the benzene ring is relatively uniform, without obvious electron-rich centers like the oxygen atom in furan. Therefore, benzene, as an electron donating group, provides relatively fewer electrons to the DPP conjugated system, and its effect on raising the HOMO energy level is not as good as furan. However, the chlorine atom in chlorobenzene has high electronegativity and exhibits an electron-withdrawing inductive effect. Meanwhile, the lone pair electrons of the chlorine atom have an electron-donating conjugative effect with the benzene ring. Nevertheless, the electron-withdrawing inductive effect generally outweighs the electron-donating conjugative effect, so chlorobenzene acts as an electron-withdrawing group. The electron-withdrawing effect reduces the electron cloud density of the DPP molecule and lowers the energy of electrons in the HOMO, thereby causing the HOMO energy level to decrease. Therefore, the HOMO energy level of DPP-Cl is the lowest among the three. It is worth noting that the LUMO does not directly accept electrons from electron donating groups, and its energy changes are mainly indirectly caused by molecular polarization or spatial effects of substituents, so its sensitivity is lower than the HOMO.

Additionally, calculations of the total electron count in the central core of DPP derivatives indicate that the total number of electrons in the acceptor follows the trend: DPP-O > DPP-Ph > DPP-Cl (Fig. 1b). The electrostatic potential distributions of the three DPP derivatives are shown in Fig. 1c–e, where warm colors represent negative charges and cool colors represent positive charges. Notably, the carbonyl group in the receptor core of DPP-O appeared the deepest red color, signifying the highest negative charge density; the carbonyl group in DPP-Cl showed light yellow, indicating the lowest negative charge density; and the carbonyl group in DPP-Ph exhibited a red color, with a negative charge density intermediate between the two. These results further confirm from the perspective of electron distribution that the electron-donating ability of furan is significantly stronger than that of benzene, while chlorobenzene has the weakest electron-donating capability.

Compared with the energy band structure of polymer PEI, as shown in Fig. 1a, it can be observed that the LUMO energy level of DPP-X (X = Cl, Ph, O) is lower than that of the polymer matrix, while the HOMO energy level is higher than that of the polymer matrix. The formation of electron traps originates from the decrease of the LUMO energy level, while hole traps are caused by the increase of the HOMO energy level (the decrease or increase is relative to the energy level of the polymer matrix).^19,28 Therefore, DPP-X (X = Cl, Ph, and O) can act as electron traps and hole traps, capturing electrons and holes. The energies of LUMO and HOMO levels correspond to electron affinity (EA) and electron ionization energy (EI), respectively. Therefore, the more the LUMO level decreases or the HOMO level increases, the higher the energy level of the introduced electron or hole traps.²⁹ That is to say, the deeper the trap, the closer the LUMO/HOMO energy level is to the Fermi level of the polymer (E_F).¹⁹ From Fig. 1a, the electron trap formed by DPP-Cl is the deepest, while the electron trap formed by DPP-O is the shallowest. In contrast, the hole traps formed by DPP-O are the deepest, while those formed by DPP-Cl are the shallowest. By comparison, the hole trap depth formed by DPP-O exceeds the electron trap depth formed by DPP-Cl, indicating that DPP-O has a stronger ability to capture charge carriers (Fig. 2a).

Fig. 2 (a) Schematic diagram of the physical cross-linking network formed by hydrogen bonding and electrostatic interactions between DPP-X (X = Cl, Ph, and O) and the PEI matrix. (b) Schematic diagram of hydrogen bonding and electrostatic non-covalent interactions between DPP-X (taking DPP-O as an example) and PEI.

In addition, the difference in electron-donating ability of DPP-X (X = Cl, Ph, and O) also has a significant impact on the non-covalent interactions (hydrogen bonding and electrostatic interactions) between DPP-X (X = Cl, Ph, and O) and PEI. The strong electron-donating groups (such as furan) increase the overall electron cloud density of DPP-O, resulting in more negative charges on the surface of the molecule. However, in the PEI chain, the phenyl group attached to the imine group usually carries a positive charge. Therefore, the stronger the electron-donating ability of DPP molecules, the stronger the electrostatic attraction between them and PEI polymers (Fig. 2a and b). In contrast, benzene has moderate electron-donating ability, while chlorobenzene has weaker electron-donating ability. Therefore, the order of the strength of the electrostatic interaction between DPP-X (X = Cl, Ph, and O) and PEI is DPP-O > DPP-Ph > DPP-Cl. The difference in electron-donating ability also affects the formation of hydrogen bonds between DPP derivatives and the PEI matrix. The furan group with strong electron-donating ability changes the electron cloud distribution of DPP-O, causing the electron cloud to concentrate more on atoms or groups with higher electronegativity or electron affinity in the conjugated system. So, the hydrogen atom connected to the DPP-O conjugated system has stronger positivity, making it easier to form hydrogen bonds with the carbonyl group (-C = O) in PEI (Fig. 2b). In summary, the stronger the electron-donating ability of the electron-donating groups in DPP derivatives, the more conducive it is for DPP-X (X = Cl, Ph, and O) to form a physical crosslinking network with PEI polymers through hydrogen bonding and electrostatic interactions. That is, a highly ordered physical cross-linked network is formed between DPP-O with strong electron-donating groups and PEI through non-covalent interactions. In contrast, the non-covalent interactions between DPP-Cl and PEI chains are the weakest, resulting in less constraint on PEI chains by DPP-Cl and the lowest orderliness of polymer chains (Fig. 2a). The differences in these non-covalent interactions and the order of molecular chains will further affect the properties of PEI-DPP-X composites, such as mechanical strength, stability, etc.

3.2. Structural characteristics

Polymer dielectric composite films of PEI-DPP-X (X = Cl, Ph, and O) were successfully prepared by the solution casting method, and the characteristic functional groups of the composite films were characterized by Fourier transform infrared spectroscopy (FTIR). As shown in Fig. S2a and b, the characteristic peaks of PEI appear in the FTIR spectra of all composites.³⁰ Notably, as depicted in Fig. S2c, the characteristic peak of the composite film at 1619 cm⁻¹ exhibits a slight shift compared to PEI (1618 cm⁻¹), which provides evidence for the formation of hydrogen-bonded non-covalent interactions between DPP-X and PEI. Furthermore, temperature-dependent FTIR spectra (Fig. S2d) show that as the temperature increases, the peak initially at 1619 cm⁻¹ gradually shifts to 1615 cm⁻¹, and the intensity of this peak decreases. This behavior indicates that the strength of hydrogen bonds weakens with increasing temperature. Meanwhile, we confirmed the existence of hydrogen bonding and electrostatic interactions between DPP-O and PEI chains through molecular dynamics simulations, as depicted in Fig. S3. Among these interactions, the electrostatic interaction energy between DPP-O and PEI (−187.726 kcal mol⁻¹) is significantly lower than the hydrogen bonding interaction energy (−8.707 kcal mol⁻¹). This result indicates that small-molecule DPP-X (X = Cl, Ph, and O) and the polymer PEI are more prone to forming physical cross-links via electrostatic interactions. UV-vis spectroscopy was further employed to analyze the structural characteristics of PEI-DPP-X (X = Cl, Ph, and O) composite films. By comparing the UV-vis spectra of PEI-DPP-X (X = Cl, Ph, and O) composite films and small molecule DPP-X (X = Cl, Ph, and O) solution (Fig. S2e and f), it was found that the characteristic absorption peak of small molecule DPP-X appeared clearly at 400–600 nm in the UV-vis spectra of the PEI-DPP-X composite films, indicating that small molecule DPP-X is stably present in the PEI matrix. Furthermore, scanning electron microscopy (SEM) characterization was performed on the cross-section of the PEI-DPP-Cl-0.2 wt% composite film, and energy-dispersive spectroscopy (EDS) was combined to analyze the characteristics of its elemental distribution. As shown in Fig. S4, no obvious pores, cracks, or other microdefects were observed inside the PEI-DPP-Cl-0.2 wt% composite film, confirming that the prepared composite film has excellent compactness. Meanwhile, the results of EDS elemental mapping analysis show that chlorine (Cl) elements are continuously and uniformly distributed in the film, indicating that small-molecule DPP-X (X = Cl, Ph, and O) has achieved uniform dispersion at the molecular level in the polymer matrix. The X-ray diffraction (XRD) patterns were used to analyze the molecular chain packing density (such as d-spacing) and order of polymer composite films. As shown in Fig. 3a, both pure PEI and PEI-DPP-Cl-0.2 wt%/PEI-DPP-Ph-0.2 wt% composite films exhibit diffused broad peaks in the low-angle region, indicating that the amorphous structure of the polymer matrix remains unchanged after the incorporation of DPP derivatives (DPP-Cl and DPP-Ph) into PEI. Notably, the 2θ values show a trend of PEI-DPP-Ph-0.2 wt% > PEI-DPP-Cl-0.2 wt% > PEI. The chain spacing (d) calculated by Bragg's equation decreases slightly with the increasing electron-donating ability of the electron-donating groups in DPP derivatives (Fig. 3b), indicating that the stronger electron-donating ability of DPP molecules is conducive to forming non-covalent interactions between DPP derivatives and PEI polymers to construct dense physical crosslinking points. Due to the presence of multiple peaks in the XRD pattern of PEI-DPP-O-0.2 wt% (Fig. S5), peak fitting analysis was performed, as shown in Fig. 3c. The results show a triple-peak pattern within the 2θ range of 10°–40°, where the broad peak at 10°–30° represents the amorphous polymer structure, the peak at 20°–30° corresponds to chain packing (ch-pack), and the peak at 20°–40° indicates π-stacking.³¹ This result confirms the existence of local ordered structures (chain packing and π-stacking) in the amorphous matrix of PEI-DPP-O-0.2 wt%. Further analysis shows that the d-spacing of the ch-pack in PEI-DPP-O-0.2 wt% is 3.64 Å, and the d-spacing of the π-stacking is 3.08 Å, with an ordered region area accounting for 44.93%. This demonstrates that when furan serves as the electron-donating group, the non-covalent interactions (H-bond interaction and electrostatic interaction) between DPP-O and the PEI polymer are strongest, thereby promoting the formation of local ordered structures.

Fig. 3 (a) XRD patterns of the pure PEI film, PEI-DPP-Cl-0.2 wt% and PEI-DPP-Ph-0.2 wt% composite films. (b) Interchain spacing determined by XRD data. (c) The resolving of multi-peak and fitting curves of XRD for PEI-DPP-O-0.2 wt%. (d) DSC curves of pure PEI and all composite films. (e) The T_g and (f) ΔC_p of pure PEI and all composite films.

After clarifying the structural characteristics of the polymer composite films, the exploration of their thermodynamic properties was further carried out. Differential scanning calorimetry (DSC) was used to determine the glass transition temperature (T_g) and specific heat capacity (ΔC_p) of all polymer films, and the results are shown in Fig. 3d–f. For pure PEI, due to its rigid molecular skeleton structure, the T_g is as high as 221.79 °C. However, the presence of hydrogen bonding and electrostatic interactions between DPP-X (X = Cl, Ph, and O) and PEI leads to a higher T_g for PEI-DPP-X composite films compared to pure PEI. As the doping ratio is only 0.2 wt%, the increase in T_g is relatively limited. Additionally, it can be observed from Fig. 3e that with the increase in the electron-donating ability of the electron-donating groups in DPP derivatives, the T_g of PEI-DPP-X composite films shows a gradient increasing trend, among which the T_g of PEI-DPP-O-0.2 wt% reaches the maximum value. This phenomenon indicates that the stronger the electron-donating ability of DPP molecules, the tighter the physical crosslinking network formed between DPP derivatives and PEI. Importantly, pure PEI has the highest ΔC_p, indicating that its molecular chains have strong mobility around the T_g. After the addition of DPP-X, the ΔC_p of the PEI-DPP-X composite film decreased, indicating that the non-covalent interactions between DPP-X and PEI (such as electrostatic interactions and hydrogen bonds) limited the thermal motion of PEI molecular chains, thereby reducing the specific heat capacity of the material.^18,32 Moreover, due to the easier formation of hydrogen bonds and electrostatic interactions between DPP-O and PEI chains, the PEI-DPP-O-0.2 wt% composite film has the highest orderliness of molecular chains, resulting in the lowest ΔC_p.

3.3. Dielectric properties

To investigate the effect of DPP derivatives with different electron-donating groups on the dielectric properties of PEI, we studied the changes in dielectric constant (ε_r) and dielectric loss (tan δ) with frequency and temperature. The variation of ε_r of all polymer films with frequency (150 Hz–10 MHz) at room temperature is shown in Fig. 4a. At 150 Hz, the ε_r value of the pure PEI film is 3.19. After doping with small molecule semiconductor DPP derivatives, the ε_r of the composite film increases with the enhancement of the electron donating ability of the DPP molecule donor group (DPP-Cl < DPP-Ph < DPP-O), but the increase is relatively small, which may be related to the small doping amount. There are two main reasons to explain this change. On the one hand, groups with strong electron-donating ability (such as furan) can enhance the dipole moment of DPP molecules, resulting in higher dipole polarization strength and thus increasing the dielectric constant. On the other hand, strong electron-donating groups (such as furan) can enhance the accumulation of interfacial charges through hydrogen bonding or electrostatic attraction, triggering stronger interfacial polarization and further increasing the dielectric constant. It is worth noting that ε_r of all polymers decreases with the increase of frequency. This is caused by the fact that, under high-frequency electric fields, the rotation of polar groups and DPP molecules cannot keep up with the change of the electric field, and the migration rate of interfacial charges is slow, making it unable to respond to high-frequency electric fields.³³ Although the dielectric loss (tan δ) of all polymers increases with the increase of frequency (Fig. 4b), they remain at a low level (<0.02).

Fig. 4 (a) The dielectric constant and (b) dielectric loss of pure PEI and all composite films as a function of frequency at room temperature. (c) Dielectric constant and (d) dielectric loss of pure PEI and all composite films at 1 kHz as a function of temperature.

Further research was conducted on the effect of temperature on ε_r and tan δ of all polymer films at 1 kHz. As shown in Fig. 4c, with the increase of temperature, ε_r of all polymer films decreases significantly. When the temperature exceeds 100 °C, the coefficient tends to stabilize, which is related to the hygroscopicity of PEI.³⁴ In addition, the PEI-DPP-X composite films exhibit extremely low tan δ at both room temperature and high temperatures, as shown in Fig. 4d. In summary, as the electron-donating ability of the electron-donating groups in DPP derivatives increases, ε_r of PEI-DPP-X composite films also increases sequentially. In addition, doping DPP derivatives into the PEI matrix not only increases ε_r of the composite film, but also exhibits significant dielectric thermal stability at high temperatures.

3.4. Breakdown strength and conductivity mechanism

The breakdown strength (E_b) of dielectric polymers and the shape factor (β) related to the scattering of the breakdown data can be analyzed using the Weibull distribution function. The higher the β value, the higher the reliability of the data.³⁵Fig. 5a–d show the Weibull distribution and breakdown strength of PEI and PEI-DPP-X (X = Cl, Ph, and O) composite films at room temperature (R.T.), 150 °C, and 200 °C, respectively. Firstly, as the temperature increases, the E_b of all polymer films gradually decreases, which is due to the increase in Joule heating and conductivity loss at high temperatures.³⁶ Secondly, whether at room temperature or high temperature, the E_b of the pure PEI film is lower than that of the PEI-DPP-X (X = Cl, Ph, and O) composite film. In addition, as the electron-donating ability increases (DPP-Cl < DPP-Ph < DPP-O), the E_b of the PEI-DPP-X (X = Cl, Ph, and O) composite film increases sequentially, with PEI-DPP-O-0.2 wt% having the highest E_b. According to the Stark–Garton equation, the breakdown strength (E_b) of a dielectric material depends on its mechanical strength, with the equation expressed as E_b = 0.606(Y/(ε₀ε_r))^0.5, where Y represents the Young's modulus. Therefore, we conducted nanoindentation tests on all polymer films, and their Young's modulus values are shown in Fig. 6a and b. The Young's modulus of all PEI-DPP-X (X = Cl, Ph, and O) composite films is higher than that of pure PEI films, and as the electron-donating ability increases, the Young's modulus of PEI-DPP-X (X = Cl, Ph, and O) composite films also increases. Among them, the Young's modulus of PEI-DPP-O-0.2 wt% reaches its maximum value (5.22 GPa). Meanwhile, we characterized the tensile strength and elongation at break of the pure PEI film and PEI-DPP-X composite films (X = Cl, Ph, and O) via tensile stress–strain tests; the relevant curves and data are shown in Fig. S6. Through comparative analysis, it can be found that as the electron-donating ability of the electron-donating groups in DPP-X (where X = Cl, Ph, and O) increases, the tensile strength and elongation at break of the corresponding composite films increase accordingly. This result is consistent with the Young's modulus test results, with PEI-DPP-O-0.2 wt% exhibiting the highest tensile strength and elongation at break. Specifically, the rigid DPP-O molecules act as “microscopic reinforcement sites”, which enhance the intermolecular bonding force of PEI, thereby increasing the Young's modulus. The strong interfacial interaction enables DPP-O and PEI to form a “synergistic stress-bearing network”. During stretching, DPP-O shares the stress and reduces the breakage of PEI chains, thus increasing the tensile strength. Additionally, DPP-O is uniformly dispersed, which does not excessively restrict the slipping and reorganization of PEI chains; meanwhile, the strong interaction allows for “flexible stress transfer”, leading to a simultaneous increase in elongation at break. In conclusion, the above test results further indicate that as the electron-donating ability of the electron-donating groups in DPP-X (where X = Cl, Ph, O) increases, the non-covalent interactions (hydrogen bonding and electrostatic interactions) between DPP-X and the PEI matrix are enhanced synchronously. The strengthening of these non-covalent interactions promotes a continuous improvement in the mechanical strength of the composite films, which leads to a corresponding increase in their breakdown strength (E_b).

Fig. 5 Weibull distribution of pure PEI and composite films at (a) room temperature, (b) 150 °C, and (c) 200 °C. (d) Bar charts of the Weibull breakdown strength of pure PEI and composite films at different temperatures.

Fig. 6 (a) The nano-indentation tests of pure PEI and composite films. (b) Young's modulus of pure PEI and composite films determined by load–displacement curves. (c) and (d) Leakage current density of pure PEI and all composite films at 150 °C and 200 °C. (e) Comparison of leakage current density between pure PEI and all composite films at high temperatures (150 °C and 200 °C). (f) TSDC curves of PEI and PEI-DPP-O-0.2 wt%.

In addition, due to the significant impact of internal charge conduction on the electrical breakdown performance of dielectric polymer thin films. Therefore, we tested the leakage current density of the pure PEI film and all composite films under the conditions of 150 °C and 200 °C. As can be seen from Fig. 6c–e, comparative analysis reveals that the stronger the electron-donating ability of the electron-donating groups in DPP-X (X = Cl, Ph, O), the smaller the leakage current of the composite films—this trend is particularly prominent under high electric field conditions. This indicates that the enhanced electron-donating ability of the electron-donating groups in DPP-X (X = Cl, Ph, and O) can lead to a corresponding improvement in the composite films’ ability to trap carriers. Among them, the PEI-DPP-O-0.2 wt% composite film exhibits the smallest leakage current, indicating its strongest charge carrier trapping ability; this result indirectly demonstrates that DPP-O, which can form deep hole traps, has a more significant advantage in charge carrier trapping. Notably, as the temperature increases, the leakage current of all polymer films increases significantly (Fig. 6e). On the one hand, thermal excitation can effectively release the bound state of charge carriers inside the dielectric polymer films, leading to an increase in the concentration of free charge carriers; on the other hand, the intensified thermal motion endows charge carriers with higher kinetic energy, significantly improving their migration efficiency. These two factors work synergistically, ultimately resulting in a substantial increase in the leakage current of the dielectric polymer films.

In the field of dielectric material research, the conduction mechanism related to the electric field and temperature involved in leakage current has always been a focus of attention. Under the conditions of 150 °C and 200 °C, the fitting results of the relationship between the leakage current density and the electric field for all polymer films are shown in Fig. 6c and d. Schottky emission belongs to the electrode-limited conduction mechanism, which depends on the electrical characteristics of the electrode–dielectric interface.³⁷ At 150 °C and low electric field (<150 MV m⁻¹), due to the different band structures of the metal electrode and dielectric polymer, the charge carriers in the metal electrode can obtain sufficient energy through thermal activation, thereby crossing the energy barrier between the metal dielectric polymer interface and increasing the conduction current of pure PEI and PEI-DPP-O-0.2 wt% composite films, which belongs to Schottky emission conduction (Fig. 6c). Among them, the high determination coefficient (R²) values obtained by fitting the conductivity curve are close to 1, indicating a very good degree of fitting. Hopping conduction occurs due to the tunneling effect of trapped electrons in the dielectrics “hopping” from one trap location to another.³⁸ As the electric field increases, the interface potential barrier significantly decreases, the carrier injection rate exceeds the trap trapping ability, and the conduction is dominated by the hopping process of the trap network inside the polymer. Therefore, Schottky emission transitions to hopping conduction. When the temperature increases to 200 °C, the thermal energy (k_BT) is much higher than the interfacial barrier, so carrier injection is hardly hindered. At the same time, some carriers gain enough energy to break free from the trap binding, and thermally activated hopping becomes the main migration mode. Hopping distance (λ) is an important parameter for hopping conduction, which can be calculated using the hopping conduction model (formula (1)).³⁹

where λ_H represents the hopping distance, v is the frequency of escaped electrons by thermal vibration at trap sites, and μ_H is the barrier height. Eqn (1) can be simplified as follows:

J(E) = A × sinh(BE) (2)

where A and B are two lumped parameters. Shorter hopping distances indicate a higher trap density, meaning that charges are more easily captured. Therefore, short-distance hopping reduces the risk of breakdown caused by carrier accumulation. Under the condition of 200 °C, the fitting results of the hopping conduction curves show that the pure PEI film has the largest hopping distance; moreover, as the electron-donating ability of the electron-donating groups in DPP-X (X = Cl, Ph, and O) increases, the hopping distance of the composite films shows a gradually decreasing trend. Among them, the PEI-DPP-O-0.2 wt% composite film has the shortest hopping distance, which indicates that this all-organic composite material has a higher trap site density. The high-density trap sites can capture injected charge carriers through strong electrostatic attraction, thereby effectively suppressing charge injection and transport under high temperature and high electric field conditions. In summary, the detailed discussion on the conductivity loss at high temperatures confirms that doping small-molecule semiconductor DPP derivatives into the PEI matrix can effectively capture charge carriers and reduce the leakage current of composite films, thereby improving their electrical breakdown performance and energy storage performance.

Given that the PEI-DPP-O-0.2 wt% composite film exhibits the lowest leakage current, to further verify that the PEI-DPP-O-0.2 wt% composite material can effectively trap charge carriers and reduce leakage current, we conducted thermally stimulated depolarization current (TSDC) tests on the pure PEI and PEI-DPP-O-0.2 wt% composite materials. As shown in Fig. 6f, the first depolarization current peak of pure PEI is at around 152 °C, while that of the PEI-DPP-O-0.2 wt% composite shifts to 191 °C, accompanied by a higher current amplitude. This indicates that the PEI-DPP-O-0.2 wt% composite forms deeper trap depths and can capture more carriers, thus more effectively restricting the conduction of internal charges under high temperature and electric fields.

3.5. Energy storage performance

The electric displacement–electric field (D–E) loops of PEI and PEI-DPP-X composite films at different temperatures are shown in Fig. S7–S9. At room temperature, when the applied electric field strength is less than 700 MV m⁻¹, the D–E loops of all samples exhibit an elongated feature; while under high-temperature conditions, when the electric field strength exceeds 500 MV m⁻¹, the D–E loops significantly “bulge”, and the residual polarization value increases obviously. This is because the charge conduction ability activated by heat at high temperatures is enhanced, enabling the trapped charges to escape from the traps more easily and participate in conduction. The discharged energy density (U_d) and the charge–discharge efficiency (η) corresponding to all polymer films at different temperatures were obtained by integrating their D–E loops. As shown in Fig. 7, the energy storage performance of the PEI-DPP-X (X = Cl, Ph, and O) polymer composite film is better than that of the pure PEI film at both room temperature and high temperatures (150 °C and 200 °C). And with the increase of the electron-donating ability of the electron-donating groups in DPP molecules (DPP-Cl < DPP-Ph < DPP-O), the energy storage performance of the composite film PEI-DPP-O-0.2 wt% is the best, followed by PEI-DPP-Ph-0.2 wt%, and the energy storage performance of PEI-DPP-Cl-0.2 wt% is only better than that of pure PEI. This trend is consistent with the variation trend of E_b. This experimental phenomenon can be explained from the following two aspects. First, the stronger the electron-donating ability (DPP-Cl < DPP-Ph < DPP-O), the greater the charge carrier trapping ability of the PEI-DPP-X (X = Cl, Ph, and O) composite films. Second, the stronger the electron-donating ability, the stronger the non-covalent interactions formed between DPP derivatives and PEI, which promotes the increase of E_b. Consequently, the dischargeable energy density increases in the order of PEI-DPP-Cl-0.2 wt% < PEI-DPP-Ph-0.2 wt% < PEI-DPP-O-0.2 wt%. Among them, when the η is 90%, the U_d values of PEI-DPP-O-0.2 wt% at room temperature, 150 °C, and 200 °C are 7.38 J cm⁻³, 5.64 J cm⁻³, and 3.98 J cm⁻³ respectively, which are 1.92 times, 1.92 times, and 2.05 times higher than those of pure PEI. It is worth noting that the U_d values of both pure PEI films and PEI-DPP-X (X = Cl, Ph, O) composite films decrease with increasing temperature, which is the result of the combined effects of intensified thermal motion of charge carriers, reduced trap capture efficiency, degradation of dielectric properties, and weakening of interfacial non-covalent interactions.

Fig. 7 Discharge energy density and discharge efficiency of pure PEI and composite films at (a) room temperature, (b) 150 °C, and (c) 200 °C. (d) Discharged energy density (η = 90%) of pure PEI and composite films at different temperatures.

3.6. Cyclic reliability and fast-discharge capability

In summary, PEI-DPP-O-0.2 wt% exhibits excellent dielectric properties, energy storage performance, mechanical properties, and thermal stability. For practical applications, its reliability and stability should also be considered. As shown in Fig. 8a, during the cyclic charge–discharge test at 200 C under an electric field of 200 MV m⁻¹, PEI-DPP-O-0.2 wt% can withstand 10⁵ cycles with the η maintained above 90%, outperforming pure PEI. This indicates that the PEI-DPP-O-0.2 wt% composite film has excellent cycling stability under high temperature and high electric fields, which is attributed to its enhanced mechanical properties and reduced electrical conductivity. Compared with batteries and supercapacitors, the fast discharge rate is one of the main advantages of dielectric capacitors. The pulsed discharge performance of commercial BOPP (120 °C), pure PEI films (200 °C), and PEI-DPP-O-0.2 wt% composite films (200 °C) was tested separately under an electric field strength of 300 MV m⁻¹. As shown in Fig. 8b, comparative analysis reveals that the PEI-DPP-O-0.2 wt% composite film exhibits the highest power density (0.38 MW cm⁻³), which is 1.65 times that of commercial BOPP at 120 °C. In addition, the discharge time (τ₉₅, defined as the time required to discharge 95% of the stored charge energy) of the PEI-DPP-O-0.2 wt% composite film is 3.87 μs, which is shorter than that of commercial BOPP (4.22 μs) and pure PEI film (4.05 μs). This indicates that the PEI-DPP-O-0.2 wt% composite film has a faster charge–discharge rate in practical applications. In addition, it was found that the U_d and η measured in different regions of the large-area PEI-DPP-O-0.2 wt% composite film (16.8 cm × 11.7 cm) remained stable (Fig. 8c), indicating excellent uniformity of the electrical properties of the composite film. Finally, the U_d of the PEI-DPP-O-0.2 wt% composite film was compared with the U_d of other recently reported dielectric composite films at 200 °C with the η of approximately 90%. As shown in Fig. 8d, the PEI-DPP-O-0.2 wt% prepared in this study exhibits advantages in energy storage.^{18,32,40–49}

Fig. 8 (a) Cyclic performance of pure PEI and composite films under a 200 MV m⁻¹ electric field at 200 °C. (b) Fast discharge capability of PEI-DPP-O-0.2 wt% (200 °C), PEI (200 °C) and BOPP (120 °C). (c) Energy storage performances of different regions in large-scale PEI-DPP-O-0.2 wt% films at different regions at 200 °C and 450 MV cm⁻¹ (the insets are the photographs of the film). (d) Comparison of discharged energy density (η = 90%) of this work and other reports at 200 °C.

4. Conclusions

In this work, we employed molecular engineering strategies to alter the chemical structure of electron donating groups in DPP derivatives and systematically studied their impact mechanisms on the carrier trap depth, interface interactions, and dielectric energy storage performance of composite materials. Among them, the stronger the electron donating ability of DPP derivative electron donating groups, the deeper the hole traps formed in the composite materials, thus effectively suppressing the increase of leakage current under high temperature and high electric field. Meanwhile, DPP molecules with strong electron donating groups (DPP-O) are more likely to form physical cross-linking networks with PEI through hydrogen bonding and electrostatic non-covalent interactions, thereby improving the mechanical strength and E_b of the composite material. At room temperature, 150 °C, and 200 °C, the E_b of PEI-DPP-O-0.2 wt% is 1.17 times, 1.11 times, and 1.14 times that of pure PEI, respectively. In addition, functional groups with strong electron-donating ability (such as furan) can enhance the dipole moment of DPP-O, thereby generating higher dipole polarization strength. Coupled with the interfacial polarization between PEI and DPP-O, which significantly improves the polarization degree of the composite film, the dielectric constant of the composite material is rendered higher than that of pure PEI. The high E_b, high ε_r, and low leakage current make the PEI-DPP-O composite film have excellent U_d. When η is 90%, the U_d of PEI-DPP-0.2 wt% reaches 7.38 J cm⁻³ at room temperature, which is 1.92 times that of pure PEI (3.85 J cm⁻³). Even at high temperatures, PEI-DPP-O-0.2 wt% can achieve excellent energy storage densities of approximately 5.64 J cm⁻³ and 3.98 J cm⁻³ at 150 °C and 200 °C, respectively, surpassing most high-temperature energy storage polymers. Finally, PEI-DPP-O-0.2 wt% can be cycled 10⁵ times at 200 °C and 200 MV m⁻¹, indicating that the composite membrane has good long-term reliability. This work precisely regulates the trap depth and interface interaction through molecular engineering, synchronously improving the electrical insulation and thermal stability of polymer composite materials, providing new ideas for solving the problems of leakage current surge and energy storage efficiency degradation in high-temperature environments.

Conflicts of interest

The authors declare no competing interests.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5tc03153b.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (52172265 and 52303312), the Science and Technology Innovation Program of Hunan Province (2022RC1074), and the State Key Laboratory of Powder Metallurgy, Central South University, Changsha, China.

References

1. R. Wang, Y. Zhu, S. Huang, J. Fu, Y. Zhou, M. Li, L. Meng, X. Zhang, J. Liang, Z. Ran, M. Yang, J. Li, X. Dong, J. Hu, J. He and Q. Li, Nat. Mater., 2025, 24, 1074–1081.
2. R. Yang, D. Ai, S. Fan, W. Zhang, X. Yang, F. Lv, Y. Cheng and X. Yu, Energy Environ. Sci., 2025, 18, 3718–3729.
3. Z. Ran, M. Yang, R. Wang, J. Li, M. Li, L. Meng, Y. Liu, J. Hu, J. He and Q. Li, ESM, 2025, 74, 103952.
4. X. Li, P. Hu, J. Jiang, J. Pan, C.-W. Nan and Y. Shen, Adv. Mater., 2025, 37, 2411507.
5. M. Yang, L. Zhou, X. Li, W. Ren and Y. Shen, Adv. Mater., 2023, 35, 2302392.
6. J. Wang, X. Ma, Y. Zhang, H. Gong, B. Peng, S. Liang, Y. Xie and H. Wang, ESM, 2025, 77, 104182.
7. Z. Chang, L. Lei, L. Zhu, Y. Quan, Z. Ren, Y. Qian, D. Dastan and Z. Shi, Mater. Horiz., 2025, 12, 284–292.
8. Q. Wang, J. Ding, W. Jiang, Z. Jiang, L. Jiang, Y. Zhou and Y. Zhang, Adv. Funct. Mater., 2025, 35, 2414616.
9. J. Xiong, M. Li, T. Lu, T. Zhang, M. Hu, M. Zhang, Y. Xie, S. Tan, H. Gong and Z. Zhang, Chem. Eng. J., 2025, 507, 160661.
10. Z. Xie, Z. Huang, H. Li, T. Xu, H. Zhao, Y. Wang, X. Pang, Z. Cao, V. Altoé and L. M. Klivansky, Energy Environ. Sci., 2025, 18, 620–630.
11. Y. Wei, L. Yang, C. Wang, Z. Zhu, Y. Dai, H. Qin and C. Xiong, Chem. Eng. J., 2025, 507, 160793.
12. N. Zhang, C. Zhang, H. Guo, J. Bai and H. Zhao, Chem. Eng. J., 2025, 508, 161058.
13. T. Zeng, L. Meng, L. Cheng, R. Wang, Z. Ran, D. Liu, J. Fu, J. He, Q. Zhou and Q. Li, Adv. Funct. Mater., 2025, 35, 2419278.
14. X. Yang, H. Luo, S. Chen, J. Peng, G. He, R. Guo and D. Zhang, Composites, Part A, 2025, 196, 109019.
15. B. Zhang, X.-m Chen, Z. Pan, P. Liu, M. Mao, K. Song, Z. Mao, R. Sun, D. Wang and S. Zhang, Adv. Funct. Mater., 2023, 33, 2210050.
16. M. Feng, Y. Feng, C. Zhang, T. Zhang, X. Tong, Q. Gao, Q. Chen and Q. Chi, Energy Environ. Mater., 2024, 7, e12571.
17. X. Du, Z. Li, T. Xu, J. Wang, S. Yan and L. Dong, Chem. Eng. J., 2020, 379, 122328.
18. B. Peng, P. Wang, H. Luo, G. He, H. Xie, Y. Liu, S. Chen, X. Li, Y. Wan and R. Guo, Mater. Horiz., 2025, 12, 1223–1233.
19. J.-W. Zha, M. Xiao, B. Wan, X. Wang, Z.-M. Dang and G. Chen, Prog. Mater. Sci., 2023, 140, 101208.
20. M. Zhang, B. Zhu, X. Zhang, S. Tan, H. Gong, X. Wei and Z. Zhang, J. Mater. Chem. A, 2022, 10, 16258–16267.
21. N. Á. Kotadiya, A. Á. Mondal, P. Á. Blom and D. Á. Andrienko, Nat. Mater., 2019, 18, 1182.
22. H. Wu, J. Wang, W. He, C. Shan, S. Fu, G. Li, Q. Zhao, W. Liu and C. Hu, Energy Environ. Sci., 2023, 16, 2274–2283.
23. Y. Zhou, Y. Zhu, W. Xu and Q. Wang, Adv. Energy Mater., 2023, 13, 2203961.
24. Q. Li, L. Chen, M. R. Gadinski, S. Zhang, G. Zhang, H. U. Li, E. Iagodkine, A. Haque, L.-Q. Chen, T. N. Jackson and Q. Wang, Nature, 2015, 523, 576–579.
25. Y. Tang, W. Xu, S. Niu, Z. Zhang, Y. Zhang and Z. Jiang, J. Mater. Chem. A, 2021, 9, 10000–10011.
26. H. Luo, X. Zhou, C. Ellingford, Y. Zhang, S. Chen, K. Zhou, D. Zhang, C. R. Bowen and C. Wan, Chem. Soc. Rev., 2019, 48, 4424–4465.
27. B. Sun, P. Hu, X. Ji, M. Fan, L. Zhou, M. Guo, S. He and Y. Shen, Small, 2022, 18, 2202421.
28. H. Yuan, Y. Zhou, Y. Zhu, S. Hu, C. Yuan, W. Song, Q. Shao, Q. Zhang, J. Hu and Q. Li, J. Phys. D: Appl. Phys., 2020, 53, 475301.
29. X. Tang, C. Ding, S. Yu, C. Zhong, H. Luo and S. Chen, Small, 2024, 20, 2306034.
30. Y. Zhang, Y. Guo, Y. Liu, Z. Shi, W. Liu, J. Su, G. Chen and D. Zhou, J. Mater. Chem. C, 2024, 12, 4426–4432.
31. Y. Jia, L. Zhai, S. Mo, Y. Liu, L.-X. Liu, X.-Y. Du, M.-H. He and L. Fan, Chin. J. Polym. Sci., 2024, 42, 1134–1146.
32. C. Yan, Y. Wan, H. Long, H. Luo, X. Liu, H. Luo and S. Chen, Adv. Funct. Mater., 2024, 34, 2312238.
33. X.-J. Liu, M.-S. Zheng, G. Chen, Z.-M. Dang and J.-W. Zha, Energy Environ. Sci., 2022, 15, 56–81.
34. Q. Li, F.-Z. Yao, Y. Liu, G. Zhang, H. Wang and Q. Wang, Annu. Rev. Mater. Res., 2018, 48, 219–243.
35. F. Liu, Q. Li, J. Cui, Z. Li, G. Yang, Y. Liu, L. Dong, C. Xiong, H. Wang and Q. Wang, Adv. Funct. Mater., 2017, 27, 1606292.
36. X. Li, H. Luo, C. Yang, F. Wang, X. Jiang, R. Guo and D. Zhang, ACS Appl. Mater. Interfaces, 2023, 15, 41828–41838.
37. Y. Shang, Y. Feng, Z. Meng, C. Zhang, T. Zhang and Q. Chi, Mater. Horiz., 2024, 11, 1528–1538.
38. X. Dong, Y. Wang, Y. Cao, N. Li, J. Fu, J. Yu and Z. Hu, Chem. Eng. J., 2024, 500, 157312.
39. X. Zhen, J. Wang, X. H. Li, Y. Jiang, Z. Li, Z.-H. Shen and X. Zhang, J. Mater. Chem. C, 2025, 13, 13952–13961.
40. S. Luo, J. Yu, T. Q. Ansari, S. Yu, P. Xu, L. Cao, H. Huang, R. Sun and C.-P. Wong, Appl. Mater. Today, 2020, 21, 100882.
41. Y. Niu, J. Dong, Y. He, X. Xu, S. Li, K. Wu, Q. Wang and H. Wang, Nano Energy, 2022, 97, 107215.
42. J. Dong, L. Li, P. Qiu, Y. Pan, Y. Niu, L. Sun, Z. Pan, Y. Liu, L. Tan and X. Xu, Adv. Mater., 2023, 35, 2211487.
43. W. Ren, M. Yang, L. Zhou, Y. Fan, S. He, J. Pan, T. Tang, Y. Xiao, C. W. Nan and Y. Shen, Adv. Mater., 2022, 34, 2207421.
44. A. A. Deshmukh, C. Wu, O. Yassin, A. Mishra, L. Chen, A. Alamri, Z. Li, J. Zhou, Z. Mutlu and M. Sotzing, Energy Environ. Sci., 2022, 15, 1307–1314.
45. H. Li, D. Ai, L. Ren, B. Yao, Z. Han, Z. Shen, J. Wang, L. Q. Chen and Q. Wang, Adv. Mater., 2019, 31, 1900875.
46. S. Cheng, Y. Zhou, Y. Li, C. Yuan, M. Yang, J. Fu, J. Hu, J. He and Q. Li, ESM, 2021, 42, 445–453.
47. J. Dong, R. Hu, X. Xu, J. Chen, Y. Niu, F. Wang, J. Hao, K. Wu, Q. Wang and H. Wang, Adv. Funct. Mater., 2021, 31, 2102644.
48. C. Yuan, Y. Zhou, Y. Zhu, J. Liang, S. Wang, S. Peng, Y. Li, S. Cheng, M. Yang and J. Hu, Nat. Commun., 2020, 11, 3919.
49. M. Liu, J. Song, H. Qin, S. Qin, Y. Zhang, W. Xia, C. Xiong and F. Liu, Adv. Funct. Mater., 2024, 34, 2313258.

---